AI Governance Institute
← News

PlanAssist Expands to 15 UK Councils, Testing Public-Sector Human Oversight at Scale

Source

Faster planning decisions as new rules and AI tackle delays

Ministry of Housing, Communities and Local Government (UK)

What happened

The UK Ministry of Housing, Communities and Local Government published Faster planning decisions as new rules and AI tackle delays on 30 October 2026. The announcement confirmed the expansion of PlanAssist to 15 English councils. The tool was developed with Google DeepMind, Google Cloud, Faculty, and participating local authorities, and trialled first at Camden, Barnet, and Dorset councils. It provides planning officers with evidence-backed initial assessments of applications, cutting time spent on key tasks by 58% and halving processing times for routine householder applications. Officers retain formal decision-making responsibility throughout. The government's stated intent is to make the tool available to every council in England by the end of 2027.

Why it matters

  • ·Accountability in AI-assisted planning decisions is not yet settled. When an AI system generates the initial assessment that an officer approves, meaningful human oversight is required. This requires documented reviewer competency and an audit trail showing officers evaluated the AI output, not just endorsed it.
  • ·The UK AI Regulation Framework places accountability with the deploying organization. Individual councils are responsible for demonstrating that their use of PlanAssist meets those expectations. This applies even though the tool is centrally procured and developed.
  • ·The Google DeepMind and Google Cloud involvement means councils are procuring AI from a frontier lab. This creates vendor due diligence obligations most local authorities have not previously needed to exercise. Contract terms, data handling, model update disclosure, and incident notification rights all require scrutiny before go-live.

Stay ahead of stories like this

Get every UK AI governance development like this one, plus the rest of the week's developments. Every Thursday.

Free every Thursday. Unsubscribe anytime.

Governance controls affected

What to do now

  • ☐Ask your legal and procurement teams to confirm that your council's PlanAssist contract includes rights to be notified of model updates, performance changes, and any incidents involving the tool before they affect live decisions.
  • ☐Review whether planning officers who use PlanAssist outputs are trained to identify errors or biased assessments in AI-generated recommendations, and document that training as evidence of meaningful human oversight.
  • ☐Establish a logging process that records, for each planning decision, that an officer reviewed the AI assessment and the basis on which they agreed or departed from it, so you can demonstrate to regulators or appellants that decisions were not simply rubber-stamped.
  • ☐Identify who within your organization owns accountability for PlanAssist governance, including escalation routes if the tool produces a materially wrong assessment that leads to a flawed decision.
  • ☐If your council is not yet in the trial but expects to join before the 2027 rollout, begin a risk classification exercise now to determine whether AI-assisted planning decisions qualify as high-risk AI use under your internal governance framework.

What to watch next

Compliance teams at UK councils should monitor whether the Ministry of Housing publishes governance standards or oversight requirements to accompany the national rollout before the end of 2027. The UK Regulations Requiring an ICO Code of Practice on AI and Automated Decision-Making (SI 2026/425) is now in force. It sets expectations for how councils document and explain AI-influenced decisions to affected parties. Broader public-sector AI deployments, including the NHS 505,000-user Copilot rollout, suggest regulators are beginning to treat scale and speed of AI adoption as a governance signal. This applies in their own right.

Related Coverage

Corporate Policy2026-11-14

FAA's $875M AI Traffic System Rolls Out Without Controller Union Input

The Federal Aviation Administration has deployed its $875 million SMART air traffic management system to all 22 air route traffic control centers across the US. This expands from an initial pilot covering Washington, DC airspace. The system uses cloud-based AI to combine weather, flight path, traffic flow, and staffing data to recommend routes and arrival sequencing. Human FAA specialists validate recommendations before they reach local controllers, who retain final authority over whether to accept them. The National Air Traffic Controllers Association, representing more than 15,000 members, stated it was not consulted in the design, testing, or implementation of SMART.

Research2026-11-12

Vendor Custody, Delegated Authority, and the Agentic AI Accountability Gap

A peer-reviewed paper published in November 2026 by Springer Nature examines how accountability is distributed when AI agents make decisions on behalf of users and organizations. The research identifies four structural risks: delegated authority without clear ownership; vendor-controlled alignment measures that deployers cannot audit; limited inspection rights over agent behavior; and gaps in incident response responsibility. The findings have direct implications for vendor risk programs, human oversight requirements, and AI governance documentation.

Research2026-11-21

METR Finds AI Excels at Coding but Falls Short on Judgment Tasks

METR, an AI evaluation nonprofit, published an updated time horizon report on November 20, 2026. It found that frontier AI models continue improving exponentially on structured tasks such as coding, doubling completable task length roughly every four months. Models fall short of human performance on open-ended, judgment-laden tasks most relevant to automating AI research and development, including designing new AI evaluations and scoring philosophical arguments. The findings have direct implications for how enterprises calibrate human oversight requirements across different AI use cases.